Porous Diagnostic Boundaries: A New Emphasis for the Bulletin

There is much discussion of the overlap between schizophrenia and bipolar disorders, and this discussion will surely be extended to include a number of disorders. This will not only be mood disorders with psychotic features, but anxiety disorders, neurodevelopmental disorders, and others. It had been hoped that Diagnostic and Statistical Manual of Mental Disorders (DSM-5) could be extensively organized by psychopathology spectrum to at least suggest which disorders group together on variables important to pathophysiology such as associated genes, neuroimaging biomarkers, and epidemiological risk factors.¹ Data is available to support some spectra such as autism spectrum including several neurodevelopmental disorders or schizophrenia spectrum including schizophrenia, schizoaffective, schizophreniform, schizotypal personality disorders, and attenuated psychoses syndrome.² However, a review of evidence for including bipolar disorder in the DSM-5 psychosis chapter with schizophrenia was not compelling.³

Similarities in dependent measures between disorders are often viewed as evidence that the diagnostic classes are not distinctive. Perhaps, bipolar and schizophrenia are different versions of a similar core pathophysiological or etiological mechanism. Or, psychoses are viewed as a continuum without separate classes validated with points of rarity. As this issue is being addressed regarding the validity of current classification, general similarities should not always be viewed as evidence of porous boundaries. The overlap between disorders needs to be substantial and relevant to a core feature or a critical validator of diagnostic class. Consider the shared and unshared genetic variance between schizophrenia and bipolar disorders. To challenge current diagnostic concepts and establish new classes, we need to know:

• Are the shared genes relevant to diagnosis class?

• Are the nonshared genes relevant to diagnostic class?

• Is overlap related to a proportion of patients in each class sharing an aspect of psychopathology that relates to the shared genes but does not relate to classification (eg, anxiety)?

And why only consider a narrow range of disorders? Consider this scenario. The schizophrenia and bipolar cases share anxiety pathology, at least a substantial proportion of patients in each class experience anxiety at a symptomatic level. It seems likely that genes that contribute to vulnerability for anxiety will be associated with both disorders giving a similarity without diagnostic significance. And there are many other disorders with anxiety symptoms, so nosology research would have little reason to interpret findings in a special relationship to schizophrenia and bipolar when these associated genes may be relevant to a number of mental disorders. This would be an instance where the target is wrong, and the question would be more robustly studied in relation to a psychopathology dimension, in this case anxiety, in several syndromes where patients often manifest pathological anxiety.

Current classification is also challenged when a point of rarity is not documented between disorders or between a disorder and the non-ill population. Here, too, care needs to be taken to assure the potentially discriminating variable is critical to diagnostic class. If one chooses reality distortion symptoms as defining schizophrenia, then a point of rarity will be difficult to demonstrate with other psychotic disorders or even the general population. This is why the prominence given to Schneiderian first rank symptoms in DSM-III and IV have been de-emphasized in DSM-5.² PANSS rating of psychosis will suggest similarity, but pressured speech associated with high energy and grandiose thoughts might be a point of rarity associated with bipolar disorder in contrast to alogia/disorganized speech in schizophrenia. If cognition is considered a core feature for schizophrenia, then it may be the developmental, prepsychotic decline in cognitive capacity that provides a point of rarity between schizophrenia and bipolar disorder. Response to antipsychotic medication would suggest a continuum across psychotic disorders, but lithium response would distinguish schizophrenia from bipolar disorder.

Section 3 of DSM-5 will provide psychopathology dimensions that may help clarify porous boundaries across psychotic syndromes.² The syndromes will continue as disorder classes, and the specific psychopathology domains will provide information more specific for each individual. The anticipated reconceptualization of mental disorders based on fundamental and differentiated etiopathophysiological knowledge remains for the future. In the meantime, we need direct comparison across diagnostic groups based on critical variables.

These issues provide an important opportunity and challenge for this journal. Schizophrenia Bulletin is an old and honored name. We intend to broaden the mission without damaging the brand. The concept schizophrenia can be deconstructed into a large number of psychopathology domains, each relevant to understanding illness, but each affecting some but not all members of the diagnostic class. These pathologies are all relevant to schizophrenia, but also to other disorders. In this regard, we would hope to be the “go to” journal for highest quality papers reporting direct comparisons between disorders within the schizophrenia spectrum and disorders outside the spectrum. For example, a report of resting mode default circuitry comparing schizotypal personality disorder, schizophrenia, bipolar with psychosis, bipolar without psychosis, major depressive disorder with and without psychosis, and non-ill volunteers. Or, studies based on the Bipolar & Schizophrenia Network on Intermediate Phenotypes (BSNIP) project that compares and contrasts endophenotypes between families with a schizophrenia proband and families with a bipolar proband.

We would have always welcomed such reports, but now wish to actively seek work across diagnostic boundaries with changes in our editorial board structure. We will add new special features devoted to brief presentations of concepts that distinguish between disorders and concepts and that define porous boundaries between schizophrenia and “nearby” disorders. The associate editors for these features will be drawn from bipolar, mood, and developmental disorders fields. We will also add editorial board members whose professional identity and experience enhance our ability to attract and review reports cutting across diagnostic boundaries. We have begun this process with an upcoming theme and will welcome proposals for themes in this area of our mission. Finally, we welcome reports based on new paradigms such as Research Diagnostic Criteria behavioral construct/neural circuit-based constructs if the potential relevance to schizophrenia-related disorders is evident.⁴